Various publications, including patents, published applications, technical articles and scholarly articles are cited throughout the specification. Each of these cited publications is incorporated by reference herein, in its entirety and for all purposes
Several disease-associated antigens are currently being targeted using therapeutic monoclonal antibodies (MAbs) because of their unique pharmacological and safety profiles. Among the disease-associated target antigens are CD20, Tumor Necrosis Factor alpha (TNF-α), Epidermal Growth Factor Receptors (EGFR), and granulocyte-macrophage colony stimulating factor.
Granulocyte macrophage-colony stimulating factor (GM-CSF) was originally discovered as a protein with the capacity to generate both granulocyte and macrophage colonies from precursor cells in mouse bone marrow, and was accordingly named (Burgess et al. (1980) Blood 56:947-58.). Subsequent studies have demonstrated a role of GM-CSF in potentiating the function of mature macrophages and granulocytes (Handman and Burgess (1979) J. Immunol. 122:1134-1137; Hamilton et al. (1980) J. Cell Physiol. 103:435-445; Gamble et al. (1985) Proc. Natl. Acad. Sci. USA 82:8667-8671), suggesting a role for GM-CSF in inflammatory responses (Hamilton et al. (1980) J. Cell Physiol. 103:435-445). As the molecule was studied it became clear that GM-CSF has other functions arising from its ability to affect the properties of more mature myeloid cells such as granulocytes, macrophages and eosinophils. The functions of GM-CSF are mediated by binding to CD116, the granulocyte-macrophage colony stimulating factor receptor, also known as colony stimulating factor 2 receptor alpha that binds GM-CSF with low affinity. The beta subunit, called CD131, which is also shared with the IL3 and IL5 receptors, has no detectable binding activity for GM-CSF by itself but is necessary for high affinity binding when in association with the alpha subunit and plays a fundamental role in signal transduction. The GM-CSF receptors are found on myeloid progenitors and mature myeloid cells including neutrophils, eosinophils, mononuclear phagocytes, and monocytes. In addition, GM-CSF receptor subunits have been shown to be present in normal, non-hematopoietic tissues such as human placenta, endothelium, and oligodendrocytes of the central nervous system.
GM-CSF plays a major biological role in the generation of granulocytes and macrophages from early bone marrow progenitors within the bone marrow. What was not appreciated at first but later uncovered was additional physiological functions of GM-CSF in host responses to external stimuli and in inflammatory and autoimmune conditions. In very early studies, GM-CSF was purified from lung tissue-conditioned medium following lipopolysaccharide (LPS) injection into mice (Burgess et al. (1977) J. Biol. Chem. 252:1998-2003). GM-CSF is considered by many investigators to be one of the major regulators of granulocyte, macrophage and eosinophil lineage cell number and activation state under normal physiological conditions. However, it has also been hypothesized that aberrant expression of GM-CSF may lead to altered immune and inflammatory responses with associated pathologic consequences. It was suggested several years ago that GM-CSF should be viewed as a proinflammatory cytokine (Hamilton et al., 1980, J. Cell Physiol. 103:435-445). Furthermore, GM-CSF may play a role in the diathesis of a multitude of human inflammatory pathologies, such as rheumatoid arthritis, autoimmune pathologies, inflammatory renal disease and inflammatory lung disorders such as asthma and chronic obstructive pulmonary disease (COPD). Interestingly, it has been proposed that there is a link between multiple sclerosis and GM-CSF (McQualter et al. (2001) J. Exp. Med., 194:873-881). In an experimental model of autoimmune encephalomyelitis, a model for multiple sclerosis, GM-CSF was found to be involved in the autoimmune-mediated demyelination.
In vivo studies following monocyte, macrophage and neutrophil treatment with GM-CSF have demonstrated that GM-CSF can activate these cell types and prolong their survival characteristics. Moreover, GM-CSF exposure results in release of inflammatory mediators from these cell types, and further studies have demonstrated the ability of these cells to kill certain organisms and even tumor cells (Hamilton (1993) Immunol. Today 14:18-24; Hamilton, (1993) Lancet 342:536-539; Takahashi, (1993) Blood 81:357-364). To determine if the in vivo studies were indicative of the function of GM-CSF in vivo, systemic administration was performed with rodents. It was shown that artificially increasing circulating levels of GM-CSF by intraperitoneal administration of the protein did result in increased numbers of both circulating neutrophils and cycling peritoneal macrophages and that there was an increase in the development and differentiation of CD5+ macrophages in the peritoneal cavity of rodents (Metcalf et al., (1987) Exp. Hematol. 15:1-9).
It has also been shown that GM-CSF can “prime” cells to respond in a more robust, synergistic manner to a second stimulus, such as LPS or interferon-gamma (Hart et al., 1988, J. Immunol. 141:1516-1521). Mice can be primed both in vitro as well as in vivo with GM-CSF so that they produce increased levels of circulating pro-inflammatory cytokines following subsequent challenge with LPS or TNF-alpha.
In a clinical setting, administration of GM-CSF into peritoneal dialysis patients resulted in a marked recruitment of macrophages (Selgas et al., 1996, Kidney Int. 50:2070-2078). Interestingly, and as predicted from the rodent studies, administration of GM-CSF in a clinical setting can result in accentuated production of inflammatory cytokines and potential unwanted side effects. For example, when patients with rheumatoid arthritis were treated with GM-CSF to correct the neutropenia associated with Felty's syndrome, their arthritis was exacerbated (Hazenberg et al., 1991, Blood 74:2769-2770). In another clinical setting, following cancer chemotherapy, GM-CSF treatment made rheumatoid arthritis worse (de Vries et al., (1991) J. Immunol. 163: 4985-4993). Systemic administration of GM-CSF to human donors increased the ability of isolated granulocytes to produce superoxide, and both accentuated the cytotoxicity of circulating monocytes as well as led to an increase in the number of monocytes (Perkins et al., 1993, Am J. Hematol. 43:279-285). Aberrant expression of GM-CSF is associated with disease of the lung in human as well. For example, it appears that upregulation of GM-CSF in the lung by minor irritants, endotoxins or infections predisposes towards TH2 immune deviation and asthma (Eisenbarth et al. (2002) J. Exp. Med. 196:1645-1651). The studies summarized above suggest that GM-CSF plays a role in the activation of the inflammatory process through cell recruitment, increased cell survival and/or priming for activation.
Several association and experimental data suggest a role for GM-CSF in asthma. The use of neutralizing antibodies in a mouse model of asthma have demonstrated the ability to suppress asthmatic phenotypes (Yamashita (2002) Cell Immunol. 219:92), while several studies measuring cytokines in BAL fluid of asthmatic patients have found an increase in GM-CSF (Gajewska (2003) Curr Drug Targets Inflamm Allergy 2:279).
Rheumatoid arthritis (RA) is a chronic inflammatory autoimmune disease for which there is ample evidence that GM-CSF may be involved. GM-CSF has been found at elevated levels in RA lesions (Xu et al. (1989) J. Clin. Invest. 83:876) and is produced in vitro by resident joint cells (chondrocytes and synovial fibroblasts) following their stimulation with inflammatory cytokines such as IL-1 and TNF-alpha (Leizer et al. (1990) Blood 76:1989). Collagen-induced arthritis (CIA) in the mouse is an autoimmune model of RA that is dependent upon both humoral and cellular immune responses to type II collagen (CII) (Seki et al. (1988) J. Immunol. 140,1477). Historically this RA phenotype is restricted to mouse strains bearing the H-2q or H-2r haplotypes and is generally performed in DBA/1 mice (Wooley (1988) Methods Enzymol. 162:361). A series of studies were performed in transgenic mice that were homozygous null for the murine GM-CSF locus (Stanley et al. (1994) Proc. Natl. Acad. Sci. USA 91:5592). Interestingly, the GM-CSF-deficient mice are resistant to the induction of collagen-induced arthritis as compared to their wild-type control litter mates (Campbell et al. (1998) J. Immunol. 161:3639-3644).
Of further interest is that GM-CSF null mice have impaired surfactant clearance that leads to murine pulmonary alveolar proteinosis (PAP), which closely mimics the human condition as described herein. Moreover, the PAP phenotype can be corrected by lung-specific delivery of the GM-CSF gene (Zsengaller et al. (1998) Hum. Gene Ther. 9:2101-2109), aerosolization of GM-CSF or bone marrow transplantation for hematopoeitic reconstitution (Reed et al. (1999) Am. J. Physiol. 276:L556-L563; Nishinakamura et al. (1996) J. Exp. Med. 183:2657-2662).
Adult human pulmonary alveolar proteinosis (PAP) is a rare disease characterized by the accumulation of phospholipids and surfactant proteins in the alveoli. It has been hypothesized that PAP is due to the inability of the alveolar macrophages and type II epithelial cells to clear excess surfactant (Mazzone et al. (2001) Clev. Clin. J. Med. 68:977-992). The diagnosis of PAP generally requires an open lung biopsy and the standard therapy for the disease is physical removal of the accumulated surfactant by whole-lung lavage (Shah et al. (2000) Thorax 55:67-77). Furthermore, patients with PAP have been shown to have circulating, neutralizing antibodies to GM-CSF, thereby implicating this cytokine as causative of the disease. Whether this autoimmune response is specific for GM-CSF is unclear. However, it has been shown that a subset of PAP patients improve with GM-CSF therapy, supporting the hypothesis that the absence of GM-CSF either by gene disruption or antibody-mediated neutralization results in the development of PAP.
There is also evidence to support a role for GM-CSF in cancer. For example, GM-CSF plays a role in the genesis and progression of leukemias, such as juvenile myelomonocytic leukemia (JMML); Emanuel P D (2004) Curr. Hematol. Rep. 3:203-209). JMML is characterized by disruption of normal hemopoiesis resulting in excessive, inappropriate proliferation of immature myeloid cells in the bone marrow. These proliferating hematopoietic cancer cells can metastasize to the spleen and liver. Interestingly, patients with JMML are hypersensitive to GM-CSF and exhibit pathologic features similar to those in transgenic mice that over-express GM-CSF (Lang et al. (1987) 51:675-86). Furthermore, GM-CSF has been shown to promote JMML cell growth and survival (Emanuel et al. (1991) Blood 77:925-9). In the transgenic mouse model of JMML, blockade of GM-CSF reduced JMML cell burden in the bone marrow, blood and spleen (Iversen et al. (1997) Blood, 90:4910-7).
It is clear from murine disease models in which GM-CSF is knocked-out as well as human diseases such as PAP in which circulating antibodies are generated in the systemic circulation to GM-CSF that this cytokine is an important mediator of pathology. Therefore an approach to develop a drug that can antagonize the activity of GM-CSF, either by developing an antibody to the cytokine itself or by blockade of the GM-CSF receptor, may be a valuable human therapeutic. Several polyclonal and monoclonal antibodies have been generated to the recombinant GM-CSF molecule. For example, Beffy et al. ((1994), Hybridoma 13:457-468), generated polyclonal antibodies to recombinant human GM-CSF in New Zealand White rabbits and monoclonal antibodies in Balb/c mice. These rabbit and some of the murine monoclonal antibodies were capable of neutralizing the activity of GM-CSF in an in vitro cell proliferation assay with MO7e cells. In further studies, Nice et al. (1990, Growth Factors 3, 159-169) epitope-mapped the binding site of one neutralizing murine anti-GM-CSF antibody, LMM102. A well-defined epitope was delineated by generating a series of digestion products from recombinant, human GM-CSF, using reverse phase HPLC fractionation to separate the fragments, followed by additional S. aureus V8 digestion of the tryptic fragments to define a product comprising two peptides linked by a disulphide bond. Three murine antibodies to human GM-CSF were generated by Dempsey et al. (1990, Hybridoma 9, 545-558) that neutralized GM-CSF in an in vitro assay system with an EC50 in the 0.1 to 1.7 nanomolar range. These antibodies did not react with murine GM-CSF or other related cytokines. All of the above-described antibodies are useful reagents for the detection of GM-CSF in human serum as well as for in vitro assays to inhibit GM-CSF signaling. However, all of these antibodies have little value as therapeutics due to the fact that they are derived from either a murine or rabbit system. Attempts have been made to generate chimeric antibodies from murine counterparts by subcloning the variable domain from the murine anti-GM-CSF antibody into a human backbone. This strategy has led to a chimeric antibody that can neutralize GM-CSF in vitro and may be useful as a therapeutic (WO 03/068924 A2).
An important aspect of a therapeutic antibody is its ability to elicit immune effector functions, such as antibody dependent cellular cytotoxicity (ADCC). Rodent MAbs, for example, have been shown to poorly mediate effector functions in humans because of sequence differences in the Fc region and therefore chimerization or humanization are required to gain optimal pharmacological properties. In addition, MAbs with fully human sequences may still fail to support ADCC if they are produced in non-human host cells that may alter native glycosylation pattern of MAbs (Shinkawa et al. (2003) J. Biol. Chem. 278:3466-73).
In view of these facts, production of therapeutic antibodies by human B-cells is preferred. Methods for generation of hybridomas secreting human MAbs have been previously reported (WO2004/046330). Therapeutic MAbs generated by human B-cells are able to exert human effector functions and have very limited immunogenicity because of their native human structure. The generation of hybridoma or Epstein-Barr virus (EBV)-transformed lymphoblastoid lines derived from human B-cells has been previously reported (Kirman et al. (2002) Hybrid Hybridomics 21:405-14; Boerner et al. (1991) J. Immunol. 147:86-95; Zafiropoulos et al. (1997) J. Immunol. Methods 200:181-90); however, information on the characterization of these antibodies and the lines with respect to their long term stability, suitability to manufacturing processes, and the antibody's pharmacological properties is limited (van Dijk et al. (2001) Curr. Opin. Chem. Biol. 5:368-74).
There is thus a need for therapeutic human antibodies for the treatment of inflammation associated with infectious, inflammatory diseases, autoimmune disorders, and other diseases such as cancer. It is further desired that such antibodies would elicit immune effector functions, as well as be well-tolerated in human patients. The present invention addresses these and other long felt needs.